What actually counts as building in public — revenue screenshots, a changelog, or narrating your day?

None of the three, as stated. The version that has ever worked for me is publishing a decision with the reasoning attached: here is what I was choosing between, here is the data I had, here is what I picked and why, and later, here is whether I was wrong. A revenue chart with no decision behind it is a number, a changelog is a receipt, and narration is a diary. All three become interesting the moment they are attached to a choice someone else is also facing, because that is what makes a stranger stop and reply. At 45 minutes a week you can write exactly one of those a fortnight and it will outperform daily narration.

I would go further than the others and say that two of your three options are not building in public at all, they are content about building. The practice, as originally meant, was that the decisions and the numbers are visible so people can learn from them and hold you to them. If you strip out the learnable part you have a marketing channel with a nicer name, which is fine, but then judge it as marketing and it usually loses to just being useful in a forum where your users already are.

I did daily narration for five months. About 150 posts, some genuinely nice conversations, zero signups I could trace, and by month four I was inventing things to have done in order to have something to post, which is the point at which it started distorting what I worked on. That is the failure mode nobody warns you about: the format quietly becomes the roadmap.

Fourteen months of doing all three, with the signups attributed as best I could. Weekly changelog: 4 signups total over the whole period, and the people who read it were already users. Monthly revenue posts: high engagement from other makers, almost no signups, and about half the replies were people selling me something. Write ups of a specific problem and how I solved it: 61 signups, of which the top two posts accounted for 40. The distinction is whether a non maker would find the post useful, and only the third category ever does.

Post the decisions, include the number that made the decision, skip the days where there is nothing. Consistency matters less than being worth reading.